General anesthesia is accompanied by decreased efficiency of oxygen exchange. This proposal will evaluate the role of hypoxic pulmonary vasoconstriction (HPV) in determining arterial oxygenation during normal and abnormal hemodynamic conditions with and without volatile and intravenous anesthetic agents. Initial studies in rats will define a unique model for rapid and economic study of the factors influencing HPV. The effects of mixed venous oxygen tension and content, and of carbon dioxide tension, pH, and temperature will be investigated. With this basis the influence of the volatile general anesthetics halothane, isoflurane and enflurane will be examined and compared with the intravenous anesthetic agents, pentothal, ketamine and fentanyl. Subsequent studies in dogs will test further the effects of anesthetic agents when the size of the hypoxic test segment is altered. From these results predictive equations will be derived including pulmonary artery pressure, cardiac output, PA02, PV02, PVC02, pHV, CV02, temperature, and anesthetic agents. All of these factors will be further defined in terms of the four physiologic determinants: test segment size, vascular smooth muscle constriction, and conductance change of the parallel circuits in normoxic and hypoxic lung regions. Further studies will investigate the responses in dogs with atalectasis or pneumonia with and without exposure to volatile anesthetics and with and without control of mixed venous oxygen tension. Finally observations will be recorded in human subjects undergoing one lung anesthesia for thoracic surgery and in patients following cardiopulmonary bypass procedures with and without pulmonary hypertension. The responses observed in these studies will be interpretated from the results of the previous investigations of variables, and the roles of anesthesia and of pulmonary hemodynamic changes in the etiology of hypoxemia will be evaluated.